CN101547171A - Channel estimation method applicable to wireless burst communication, and system thereof - Google Patents

Abstract

The invention discloses a channel estimation method applicable to wireless burst communication, and a system thereof. The method comprises the following steps: emitting a signal frame consisting of a pseudo-random sequence frame head with a cyclic prefix and a frame body in a burst form; performing clock recovery and carrier recovery on the signal frame; performing relevant operation on the signal frame to find an initial position of a pseudo-random sequence; and utilizing the received pseudo-random sequence to perform channel estimation so as to obtain channel parameters. The invention realizes unbiased estimation of the channel parameters by properly designing the structure of the frame and utilizing the pseudo-random sequence with the cyclic prefix, can realize the multiplication and inversion of matrixes only through addition and shift operation by properly selecting channel length, and is simple in operation and easy to realize on hardware.

Description

A kind of channel estimation methods and system thereof that is applicable to wireless burst communication

Technical field

The present invention relates to digital information transmission technical field, particularly relate to a kind of channel estimation methods and system thereof that is applicable to wireless burst communication.

Background technology

The multidiameter fading channel that the wireless burst communication channel becomes normally the time for guaranteeing the accuracy of communication, need offset the adverse effect of decline by equilibrium, and channel estimating is the basis of channel equalization.Divide from method, channel estimating mainly contains based on pilot tone auxiliary algorithm for estimating and blind algorithm for estimating, and is wherein simple because of method based on the auxiliary algorithm of pilot tone, and the estimated accuracy height has obtained using widely.In the algorithm auxiliary based on pilot tone, a kind of method commonly used is to send training sequence in time domain, utilize maximum likelihood estimate to estimate channel parameter, wherein obtained using widely as training sequence with PN (Pseudo-random Noise, pseudo noise) sequence.In the standard GB 200600-2006 " digital television ground broadcast transmission system frame structure, chnnel coding and modulation " of China's in August, 2006 promulgation, promptly in frame head, use the PN sequence to reach channel estimating synchronously.But existing channel estimation methods mainly proposes at the continuous communiction pattern, record is less at the document of the channel estimation methods that the burst communication channel proposes, and need in the channel estimating matrix inversion, on hardware, be difficult for realizing that solution commonly used now is to try to achieve approximate solution with method for simplifying.Therefore the channel estimation methods at the wireless burst communication channel can not satisfy the requirement that computing is simple, precision is high simultaneously in the prior art, and is difficult for realizing on hardware.

Summary of the invention

The purpose of this invention is to provide and a kind ofly be applicable to that frame structure wireless burst communication, that utilize a kind of suitable design carries out the method and the system thereof of channel estimating.

For achieving the above object, technical scheme of the present invention provides a kind of channel estimation methods that is applicable to wireless burst communication, may further comprise the steps:

S1 utilizes emitter to be launched with burst form by the signal frame that frame head and frame are formed, and enters Channel Transmission; Described frame head comprises that length is the pseudo random sequence x=(x of N ₀, x ₁..., x _N-1) ^T, the data of described frame for needing to send, wherein T represents transposition;

S2 utilizes the channel estimating unit in the receiving system to carry out channel estimating according to received pseudo random sequence, obtains channel parameter.

Described pseudo random sequence is the m sequence.Described frame head comprises that also length is the Cyclic Prefix of M, and the frame head that described pseudo random sequence adds the above Cyclic Prefix is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, x wherein _i=1+j or x _i=-1-j, i=1,2 ..., N-1.When utilizing described emitter to launch described signal frame, adopt the quarternary phase-shift keying (QPSK) modulation system that described signal frame is modulated, wherein the frame head of the signal frame of I, Q two-way is identical.

Above-mentioned steps S2 is specially:

Utilize channel estimating unit in the receiving system according to described pseudo random sequence calculating channel parameter:

Described pseudo random sequence expression formula is y=(y ₀, y ₁..., y _N-1) ^T, and satisfy y=Ah+w, wherein y _i=c+jd, i=1,2 ..., N-1, c, d are respectively the pseudo random sequence data of I, the reception of Q two-way, channel parameter h=(h ₀, h ₁..., h _L-1) ^T, L is the length of channel, w is that length is the white Gaussian noise of N, and

Its first row promptly are vector x, and other each row are cyclic shifts of first row.W is an additive white Gaussian noise, and length is N.

At first solve h=(A according to maximum likelihood estimate ^HA) ^-1A ^HY, wherein H represents conjugate transpose, definition z=(z ₀, z ₁..., z _L-1 ^T=A ^HY,

Because matrix A ^HEvery or be 1-j, perhaps be-1+j, for the i item y that receives vector y _i=c+jd, i=1,2 ..., N-1, and

A ^HIn each and y each long-pending for (1-j) (c+jd)=(c+d)+j (c+d) or

(-1+j) (c+jd)=(c-d)+and j (c-d), try to achieve every value of z by addition;

According to the relevant nature of m sequence, when phase deviation was 0, its correlation was N again, otherwise its correlation is-1, obtains

$A^{H}A={\left[\begin{array}{cccc}2N& -2& \·\·\·& -2\\ -2& 2N& \·\·\·& -2\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ -2& -2& \·\·\·& 2N\end{array}\right]}_{L\×L}$

Here coefficient 2 occurring is because every value of x is the real part plural number identical with imaginary part.

To A ^HA inverts, and obtains

${\left(A^{H}A\right)}^{-1}=\frac{1}{2(N+1)(N-L+1)}\left[\begin{array}{cccc}N-L+2& 1& \·\·\·& 1\\ 1& N-L+2& \·\·\·& 1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ 1& 1& \·\·\·& N-L+2\end{array}\right]$

$=\frac{1}{2(N+1)(N-L+1)}(\left[1\right]+(N-L+1)I)$

Wherein [1] expression all elements is 1 matrix entirely, I representation unit battle array.

Definition $a=\underset{i=1}{\overset{L-1}{\mathrm{\Σ}}}{z}_i,$ q＝(q ₁，q ₂，…，q ^L-1) ^T，

According to q _i=a+ (N-L+1) z _i, i=1,2 ..., L-1 tries to achieve every value of q;

Last basis

$h=\frac{1}{2(N+1)(N-L+1)}(\left[\begin{array}{c}a\\ a\\ \·\\ \·\\ \·\\ a\end{array}\right]+(N-L+1)z)$

$=\frac{1}{2(N+1)(N-L+1)}q$

Obtain every value of channel parameter.

Wherein, between described step S1 and step S2, also comprise step S11: utilize clock and carrier recovery unit in the receiving system that the described signal frame in the channel is carried out clock recovery and carrier wave recovery.Between described step S11 and step S2, also comprise step S12: utilize the arithmetic element in the receiving system that the signal frame that carries out after clock recovery and carrier wave recover is carried out related operation, to find the original position of described pseudo random sequence, then described signal frame is sent to the channel estimating unit in the receiving system.The length M of described Cyclic Prefix is more than or equal to the length L of channel, and the value that satisfies N-L+1 is 2 integer power, so that multiplying each other and invert by addition and shift operation realization matrix only.Here need to prove, when the value that does not satisfy N-L+1 is 2 integer power, also can realize above-mentioned purpose, but calculation of complex.The duration of described burst communication is enough short, so that utilize channel information that channel estimating obtains effective in during burst communication.

Technical scheme of the present invention also provides a kind of channel estimation system that is applicable to wireless burst communication, comprising:

Emitter is used for the signal frame of being made up of frame head and frame is launched with burst form, enters Channel Transmission; Described frame head comprises that length is the pseudo random sequence x=(x of N ₀, x ₁..., x _N-1) ^T, the data of described frame for needing to send, wherein T represents transposition; With

Receiving system comprises channel estimating unit, and described channel estimating unit is used to utilize received pseudo random sequence to carry out channel estimating, obtains channel parameter.

Described pseudo random sequence is the m sequence.Described frame head comprises that also length is the Cyclic Prefix of M, and the frame head that described pseudo random sequence adds the above Cyclic Prefix is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, x wherein _i=1+j or x _i=-1-j, i=1,2 ..., N-1.When utilizing described emitter to launch described signal frame, adopt the quarternary phase-shift keying (QPSK) modulation system that described signal frame is modulated, wherein the frame head of the signal frame of I, Q two-way is identical.

Described receiving system comprises:

Channel estimating unit is used for according to the pseudo random sequence calculating channel parameter that receives,

The expression formula of described pseudo random sequence is y=(y ₀, y ₁..., y _N-1) ^T, and satisfy y=Ah+w, wherein y _i=c+jd, i=1,2 ..., N-1, c, d are respectively the pseudo random sequence data of I, the reception of Q two-way, channel parameter h=(h ₀, h ₁..., h _L-1) ^T, L is the length of channel, w is that length is the white Gaussian noise of N, and

Its first row promptly are vector x, and other each row are cyclic shifts of first row.W is an additive white Gaussian noise, and length is N; With

Described channel estimating unit is at first according to solving h=(A by maximum likelihood estimate ^HA) ^-1A ^HY, wherein H represents conjugate transpose, definition z=(z ₀, z ₁..., z _L-1) ^T=A ^HY is because matrix A ^HEvery or be 1-j, perhaps be-1+j, for the i item y that receives vector y _i=c+jd, i=1,2 ..., N-1, and

A ^HIn each and y each long-pending for (1-j) (c+jd)=(c+d)+j (c+d) or

(-1+j) (c+jd)=(c-d)+and j (c-d), try to achieve every value of z by addition;

Character according to the m sequence obtains again

$A^{H}A={\left[\begin{array}{cccc}2N& -2& \·\·\·& -2\\ -2& 2N& \·\·\·& -2\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ -2& -2& \·\·\·& 2N\end{array}\right]}_{L\×L}$

Here coefficient 2 occurring is because every value of x is the real part plural number identical with imaginary part.

To A ^HA inverts, and obtains

${\left(A^{H}A\right)}^{-1}=\frac{1}{2(N+1)(N-L+1)}\left[\begin{array}{cccc}N-L+2& 1& \·\·\·& 1\\ 1& N-L+2& \·\·\·& 1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ 1& 1& \·\·\·& N-L+2\end{array}\right]$

$=\frac{1}{2(N+1)(N-L+1)}(\left[1\right]+(N-L+1)I)$

Wherein [1] expression all elements is 1 matrix entirely, I representation unit battle array.

Definition $a=\underset{i=1}{\overset{L-1}{\mathrm{\Σ}}}{z}_i,$ q＝(q ₁，q ₂，…，q _L-1) ^T，

According to q _i=a+ (N-L+I) z _i, i=1,2 ..., L-1 tries to achieve every value of q;

Last basis

$h=\frac{1}{2(N+1)(N-L+1)}(\left[\begin{array}{c}a\\ a\\ \·\\ \·\\ \·\\ a\end{array}\right]+(N-L+1)z)$

$=\frac{1}{2(N+1)(N-L+1)}q$

Obtain every value of channel parameter.

Described receiving system also comprises clock and carrier recovery unit, and described clock and carrier recovery unit are used for the described signal frame of channel is carried out clock recovery and carrier wave recovery.Described receiving system also comprises arithmetic element, described arithmetic element is used for the signal frame that carries out after clock recovery and carrier wave recover is carried out related operation, to find the original position of described pseudo random sequence, then described signal frame is sent to the channel estimating unit in the receiving system.The length M of described Cyclic Prefix is more than or equal to the length L of channel, and the value that satisfies N-L+1 is 2 integer power, so that multiplying each other and invert by addition and shift operation realization matrix only.Here need to prove, when the value that does not satisfy N-L+1 is 2 integer power, also can realize above-mentioned purpose, but calculation of complex.The duration of described burst communication is enough short, so that utilize channel information that channel estimating obtains effective in during burst communication.

Technique scheme has following advantage: utilize the PN training sequence that has Cyclic Prefix to realize the nothing of channel parameter is estimated partially; By suitable selective channel length, only by addition and displacement can realization matrix multiply each other, inversion operation, computing is simple, can realize not having estimation partially on hardware.

Description of drawings

Fig. 1 is the signal frame structure schematic diagram that is used to carry out channel estimating of the embodiment of the invention;

Fig. 2 is the channel estimation methods flow chart that is applicable to wireless burst communication of the embodiment of the invention;

Fig. 3 is 9 grades of m sequencer structural representations of the embodiment of the invention;

Fig. 4 is that the hardware implementation phase of the embodiment of the invention shifts schematic diagram;

Fig. 5 is the channel estimation system structure chart that is applicable to wireless burst communication of the embodiment of the invention.

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used to illustrate the present invention, but are not used for limiting the scope of the invention.

As shown in Figure 2, the channel estimation methods flow chart that is applicable to wireless burst communication for the embodiment of the invention may further comprise the steps:

Step 201 utilizes emitter to be launched with burst form by the signal frame that frame head and frame are formed, and enters Channel Transmission; Described frame head comprises that length is the pseudo random sequence x=(x of N=511 ₀, x ₁..., x _N-1) ^T, the data of described frame for needing to send, wherein T represents transposition;

Step 202, utilize clock in the receiving system and carrier recovery unit in the channel signal frame carry out clock recovery and carrier wave recovers.

Step 203 utilizes the arithmetic element in the receiving system that the signal frame that carries out after clock recovery and carrier wave recover is carried out related operation, to find the original position of described pseudo random sequence, then described signal frame is sent to the channel estimating unit in the receiving system.

Step 204 utilizes the channel estimating unit in the receiving system to carry out channel estimating according to received pseudo random sequence, obtains channel parameter.

Described pseudo random sequence is the m sequence.Described frame head comprises that also length is the Cyclic Prefix of M=300, and the frame head that described pseudo random sequence adds the above Cyclic Prefix is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, x wherein _i=1+j or x _i=-1-j, i=1,2 ..., N-1.When utilizing emitter to launch described signal frame, adopt the quarternary phase-shift keying (QPSK) modulation system that described signal frame is modulated, wherein the frame head of the signal frame of I, Q two-way is identical.

Above-mentioned steps 204 is specially: utilize channel estimating unit in the receiving system according to described pseudo random sequence calculating channel parameter:

Described pseudo random sequence expression formula is y=(y ₀, y ₁..., y _N-1) ^T, and satisfy y=Ah+w, wherein y _i=c+jd, i=1,2 ..., N-1, c, d are respectively the pseudo random sequence data of I, the reception of Q two-way, channel parameter h=(h ₀, h ₁..., h _L-1) ^T, L is the length of channel, L=256, w are that length is the white Gaussian noise of N=511, and

Its first row promptly are vector x, and other each row are cyclic shifts of first row.W is an additive white Gaussian noise, and length is N.

At first solve h=(A according to maximum likelihood estimate ^HA) ^-1A ^HY, wherein H represents conjugate transpose, definition z=(z ₀, z ₁..., z _L-1) ^T=A ^HY,

Because matrix A ^HEvery or be 1-j, perhaps be-1+j, for the i item y that receives vector y _i=c+jd, i=1,2 ..., N-1, and

A ^HIn each and y each long-pending for (1-j) (c+jd)=(c+d)+j (c+d) or

(-1+j) (c+jd)=(c-d)+and j (c-d), try to achieve every value of z by addition;

According to the relevant nature of m sequence, when phase deviation was 0, its correlation was N again, otherwise its correlation is-1, obtains

$A^{H}A={\left[\begin{array}{cccc}2N& -2& \·\·\·& -2\\ -2& 2N& \·\·\·& -2\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ -2& -2& \·\·\·& 2N\end{array}\right]}_{L\×L}$

Here coefficient 2 occurring is because every value of x is the real part plural number identical with imaginary part.

To A ^HA inverts, and obtains

${\left(A^{H}A\right)}^{-1}=\frac{1}{2(N+1)(N-L+1)}\left[\begin{array}{cccc}N-L+2& 1& \·\·\·& 1\\ 1& N-L+2& \·\·\·& 1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ 1& 1& \·\·\·& N-L+2\end{array}\right]$

$=\frac{1}{2(N+1)(N-L+1)}(\left[1\right]+(N-L+1)I)$

Wherein [1] expression all elements is 1 matrix entirely, I representation unit battle array.

Definition $a=\underset{i=1}{\overset{L-1}{\mathrm{\Σ}}}{z}_i,$ q＝(q ₁，q ₂，…，q _L-1) ^T，

According to q _i=a+ (N-L+1) z _i, i=1,2 ..., L-1 tries to achieve every value of q;

Last basis

$h=\frac{1}{2(N+1)(N-L+1)}(\left[\begin{array}{c}a\\ a\\ \·\\ \·\\ \·\\ a\end{array}\right]+(N-L+1)z)$

$=\frac{1}{2(N+1)(N-L+1)}q$

Obtain every value of channel parameter.

Because N+1=512=2 ⁹, N-L+1=256=2 ⁸, so the parameters among q and the h only can be tried to achieve by addition and displacement.In the h parameters of finally trying to achieve, back 18 is fractional part, and everybody is integer part for the front, can accept or reject according to the precision needs.

The duration of above-mentioned burst communication is enough short, so that utilize channel information that channel estimating obtains effective in during burst communication.

The signal frame structure schematic diagram that is used to carry out channel estimating in the said method comprises frame head 101 and frame 102 as shown in Figure 1, and described frame head 101 comprises that length is pseudorandom (Pseudo-random Noise, PN) the sequence 104:x=(x of N=511 ₀, x ₁..., x _N-1) ^T, the data of described frame 102 for needing to send, wherein T represents transposition.Described frame head 101 comprises that also length is the Cyclic Prefix 103 of M=300.Described PN sequence 104 is 9 grades of m sequences, and 9 grades of m sequencer structures of present embodiment as shown in Figure 3.The frame head 101 that the PN sequence adds the above Cyclic Prefix 103 is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, x _i=1+j or x _i=-1-j, i=1,2 ..., N-1.

The channel estimation system structure chart that is applicable to wireless burst communication of the embodiment of the invention comprises as shown in Figure 5:

Emitter 501 is used for the signal frame of being made up of frame head and frame is launched with burst form, enters Channel Transmission; Described frame head comprises that length is the pseudo random sequence x=(x of N ₀, x ₁..., x _N-1) ^T, the data of described frame for needing to send, wherein T represents transposition; With

Receiving system 502 comprises channel estimating unit 505, and described channel estimating unit 505 is used to utilize received pseudo random sequence to carry out channel estimating, obtains channel parameter.

Described pseudo random sequence is the m sequence.Described frame head comprises that also length is the Cyclic Prefix of M=300, and the frame head that described pseudo random sequence adds the above Cyclic Prefix is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, x wherein _i=1+j or x _i=-1-j, i=1,2 ..., N-1.When utilizing the described signal frame of emitter 501 emission, adopt the quarternary phase-shift keying (QPSK) modulation system that described signal frame is modulated, wherein the frame head of the signal frame of I, Q two-way is identical.

The expression formula of described pseudo random sequence is y=(y ₀, y ₁..., y _N-1) ^T, and satisfy y=Ah+w, wherein y _i=c+jd, i=1,2 ..., N-1, c, d are respectively the pseudo random sequence data of I, the reception of Q two-way, channel parameter h=(h ₀, h ₁..., h _L-1) ^T, L is the length of channel, L=256, w are that length is the white Gaussian noise of N=511, and

Its first row promptly are vector x, and other each row are cyclic shifts of first row.W is an additive white Gaussian noise, and length is N; With

Described channel estimating unit 505 is at first according to solving h=(A by maximum likelihood estimate ^HA) ^-1A ^HY, wherein H represents conjugate transpose, definition z=(z ₀, z ₁..., z _L-1 ^T=A ^HY is because matrix A ^HEvery or be 1-j, perhaps be-1+j, for the i item y that receives vector y _i=c+jd, i=1,2 ..., N-1, and

A ^HIn each and y each long-pending for (1-j) (c+jd)=(c+d)+j (c+d) or

(-1+j) (c+jd)=(c-d)+and j (c-d), try to achieve every value of z by addition;

Character according to the m sequence obtains again

$A^{H}A={\left[\begin{array}{cccc}2N& -2& \·\·\·& -2\\ -2& 2N& \·\·\·& -2\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ -2& -2& \·\·\·& 2N\end{array}\right]}_{L\×L}$

Here coefficient 2 occurring is because every value of x is the real part plural number identical with imaginary part.

To A ^HA inverts, and obtains

${\left(A^{H}A\right)}^{-1}=\frac{1}{2(N+1)(N-L+1)}\left[\begin{array}{cccc}N-L+2& 1& \·\·\·& 1\\ 1& N-L+2& \·\·\·& 1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ 1& 1& \·\·\·& N-L+2\end{array}\right]$

$=\frac{1}{2(N+1)(N-L+1)}(\left[1\right]+(N-L+1)I)$

Wherein [1] expression all elements is 1 matrix entirely, I representation unit battle array.

Definition $a=\underset{i=1}{\overset{L-1}{\mathrm{\Σ}}}{z}_i,$ q＝(q ₁，q ₂，…，q _L-1) ^T，

According to q _i=a+ (N-L+1) z _i, i=1,2 ..., L-1 tries to achieve every value of q;

Last basis

$h=\frac{1}{2(N+1)(N-L+1)}(\left[\begin{array}{c}a\\ a\\ \·\\ \·\\ \·\\ a\end{array}\right]+(N-L+1)z)$

$=\frac{1}{2(N+1)(N-L+1)}q$

Obtain every value of channel parameter.

Described receiving system 502 also comprises clock and carrier recovery unit 503, and described clock and carrier recovery unit 503 are used for the described signal frame of channel is carried out clock recovery and carrier wave recovery.Described receiving system 502 also comprises arithmetic element 504: described arithmetic element 504 is used for the signal frame that carries out after clock recovery and carrier wave recover is carried out related operation, to find the original position of described pseudo random sequence, then described signal frame is sent to the channel estimating unit 505 in the receiving system.The duration of above-mentioned burst communication is enough short, so that utilize channel information that channel estimating obtains effective in during burst communication.

Because N+1=512=2 ⁹, N-L+1=256=2 ⁸, so the parameters among q and the h only can be tried to achieve by addition and displacement.In the h parameters of finally trying to achieve, back 18 is fractional part, and everybody is integer part for the front, can accept or reject according to the precision needs.

The signal frame structure schematic diagram that is used to carry out channel estimating in the said method comprises frame head 101 and frame 102 as shown in Figure 1, and described frame head 101 comprises that length is pseudorandom (Pseudo-random Noise, PN) the sequence 104:x=(x of N=511 ₀, x ₁..., x _N-1) ^T, the data of described frame 102 for needing to send, wherein T represents transposition.Described frame head 101 comprises that also length is the Cyclic Prefix 103 of M=300.Described PN sequence 104 is 9 grades of m sequences, and 9 grades of m sequencer structures of present embodiment as shown in Figure 3.The frame head 101 that the PN sequence adds the above Cyclic Prefix 103 is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, x _i=1+j or x _i=-1-j, i=1,2 ..., N-1.

Describe below and how to go up the realization channel estimating at FPGA (Field Programmable Gate Array, element programmable gate array).

The hardware resource that need use has:

Registers group 1, length is 511, is used to deposit vector x and displacement thereof.Because vector x is the real part plural number identical with imaginary part, each of x represents with 1 bit that in register 1 represents 1+j, 0 representative-1-j;

Registers group 2, length are 256, and its 1～No. 256 register is label 1,511,510 in corresponding registers group 1 respectively ..., 257 register, each row of representing matrix A successively behind registers group 1 ring shift left;

Registers group 3, length is 256, is used to deposit vector z=A ^HThe result of calculation of y;

Register 4 is used to deposit scalar a;

Register 5 is used to deposit the result of calculation of vector q.

Realize channel estimating through four-stage on FPGA, i.e. in the stage 1～4, its phase transition relation specifically describes as follows as shown in Figure 4:

Stage 1: wait for the arrival of burst frame, put registers group 1 and be initial condition, registers group 3 and register 4 clear 0.

Stage 2: the burst frame arrives, and the phase is imported a y of the PN sequences y that receives weekly _iRegisters group 1 ring shift left, then 1～L of corresponding matrix A is capable successively for registers group 2.Registers group 3 is 1 or 0 according to the value of registers group 2, with each register value and (1-j) y _iOr (1+j) y _iAddition obtains new register value, and through 511 all after dates, each register value of registers group 3 is every value of vector z.

Stage 3: registers group 3 ring shift lefts, register 4 obtains new register value with first register value addition of its register value and registers group 3.Through 256 all after dates, the value of register 4 is scalar a.It is identical when the value of each register began with the stages 3 in the registers group 3.

Stage 4: registers group 3 ring shift lefts, the move to left result of addition after 8 of register 5 first register values for the value of register 4 and registers group 3, be followed successively by the every of vector q, thereafter 18 is fractional part, everybody is integer part for the front, promptly can be used as channel estimation value output after suitably accepting or rejecting.

As can be seen from the above embodiments, the embodiment of the invention has realized the nothing of channel parameter is estimated partially by the PN training sequence that employing has Cyclic Prefix; By suitable selective channel length, only by addition and displacement can realization matrix multiply each other, inversion operation, computing is simple, can realize not having estimation partially on hardware.

The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the technology of the present invention principle; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.

Claims (18)

1, a kind of channel estimation methods that is applicable to wireless burst communication is characterized in that, may further comprise the steps:

S1 utilizes emitter to be launched with burst form by the signal frame that frame head and frame are formed, and enters Channel Transmission; Described frame head comprises that length is the pseudo random sequence x=(x of N ₀, x ₁..., x _N-1) ^T, the data of described frame for needing to send, wherein T represents transposition;

S2 utilizes the channel estimating unit in the receiving system to carry out channel estimating according to received pseudo random sequence, obtains channel parameter.

2, channel estimation methods as claimed in claim 1 is characterized in that, described pseudo random sequence is the m sequence.

3, channel estimation methods as claimed in claim 2 is characterized in that, described frame head comprises that also length is the Cyclic Prefix of M, and the frame head that described pseudo random sequence adds the above Cyclic Prefix is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, wherein xi=1+j or xi=-1-j, i=1,2 ..., N-1.

4, channel estimation methods as claimed in claim 3 is characterized in that, when utilizing described emitter to launch described signal frame, adopts the quarternary phase-shift keying (QPSK) modulation system that described signal frame is modulated, and wherein the frame head of the signal frame of I, Q two-way is identical.

5, channel estimation methods as claimed in claim 4 is characterized in that, described step S2 is specially:

Utilize channel estimating unit in the receiving system according to described pseudo random sequence calculating channel parameter:

Described pseudo random sequence expression formula is y=(y ₀, y ₁..., y _N-1) ^T, and satisfy y=Ah+w, wherein y _i=c+jd, i=1,2 ..., N-1, c, d are respectively the pseudo random sequence data of I, the reception of Q two-way, channel parameter h=(h ₀, h ₁..., h _L-1) ^T, L is the length of channel, w is that length is the white Gaussian noise of N, and

At first solve h=(A according to maximum likelihood estimate ^HA) ^-1A ^HY, wherein H represents conjugate transpose, definition z=(z ₀, z ₁..., z _L-1) ^T=A ^HY tries to achieve every value of z by addition;

Character according to the m sequence obtains again

$A^{H}A={\left[\begin{array}{cccc}2N& -2& \·\·\·& -2\\ -2& 2N& \·\·\·& -2\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ -2& -2& \·\·\·& 2N\end{array}\right]}_{L\×L}$

To A ^HA inverts, and obtains

${\left(A^{H}A\right)}^{-1}=\frac{1}{2(N+1)(N-L+1)}\left[\begin{array}{cccc}N-L+2& 1& \·\·\·& 1\\ 1& N-L+2& \·\·\·& 1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ 1& 1& \·\·\·& N-L+2\end{array}\right]$

$=\frac{1}{2(N+1)(N-L+1)}(\left[1\right]+(N-L+1)I)$

Wherein [1] expression all elements is 1 matrix entirely, I representation unit battle array.

Definition $a=\underset{i=1}{\overset{L-1}{\mathrm{\Σ}}}{z}_i,q={(q_1,q_2,\·\·\·,q_{L-1})}^T,$

According to q _i=a+ (N-L+1) z _i, i=1,2 ..., L-1 tries to achieve every value of q;

Last basis

$h=\frac{1}{2(N+1)(N-L+1)}(\left[\begin{array}{c}a\\ a\\ \·\\ \·\\ \·\\ a\end{array}\right]+(N-L+1)z)$

$=\frac{1}{2(N+1)(N-L+1)}q$

Obtain every value of channel parameter.

6, channel estimation methods as claimed in claim 4, it is characterized in that, between described step S1 and step S2, also comprise step S11: utilize clock and carrier recovery unit in the receiving system that the described signal frame in the channel is carried out clock recovery and carrier wave recovery.

7, channel estimation methods as claimed in claim 4, it is characterized in that, between described step S11 and step S2, also comprise step S12: utilize the related operation unit in the receiving system that the signal frame that carries out after clock recovery and carrier wave recover is carried out related operation, to find the original position of described pseudo random sequence, then described signal frame is sent to the channel estimating unit in the receiving system.

8, channel estimation methods as claimed in claim 5, it is characterized in that, the length M of described Cyclic Prefix is more than or equal to the length L of channel, and the value that satisfies N-L+1 is 2 integer power, so that multiplying each other and invert by addition and shift operation realization matrix only.

9, channel estimation methods as claimed in claim 4 is characterized in that, the duration of described burst communication is enough short, so that utilize channel information that channel estimating obtains effective in during burst communication.

10, a kind of channel estimation system that is applicable to wireless burst communication is characterized in that, comprising:

Emitter is used for the signal frame of being made up of frame head and frame is launched with burst form, enters Channel Transmission; Described frame head comprises that length is the pseudo random sequence x=(x of N ₀, x ₁..., x _N-1) ^T, the data of described frame for needing to send, wherein T represents transposition; With

Receiving system comprises channel estimating unit, and described channel estimating unit is used to utilize received described pseudo random sequence to carry out channel estimating, obtains channel parameter.

11, channel estimation system as claimed in claim 10 is characterized in that, described pseudo random sequence is the m sequence.

12, channel estimation system as claimed in claim 11 is characterized in that, described frame head comprises that also length is the Cyclic Prefix of M, and the frame head that described pseudo random sequence adds the above Cyclic Prefix is (x _N-M..., x _N-1, x ₀, x ₁..., x _N-1) ^T, x wherein _i=1+j or x _i=-1-j, i=1,2 ..., N-1.

13, channel estimation system as claimed in claim 12 is characterized in that, when utilizing described emitter to launch described signal frame, adopts the quarternary phase-shift keying (QPSK) modulation system that described signal frame is modulated, and wherein the frame head of the signal frame of I, Q two-way is identical.

14, channel estimation system as claimed in claim 13 is characterized in that, described receiving system comprises:

Channel estimating unit is used for according to the described pseudo random sequence calculating channel parameter that receives,

The expression formula of the described pseudo random sequence that receives is y=(y ₀, y ₁..., y _N-1) ^T, and satisfy y=Ah+w, wherein y _i=c+jd, i=1,2 ..., N-1, c, d are respectively the pseudo random sequence data of I, the reception of Q two-way, channel parameter h=(h ₀, h ₁..., h _L-1) ^T, L is the length of channel, w is that length is the white Gaussian noise of N, and

Described channel estimating unit is at first according to solving h=(A by maximum likelihood estimate ^HA) ^-1A ^HY, wherein H represents conjugate transpose, definition z=(z ₀, z ₁..., z _L-1) ^T=A ^HY tries to achieve every value of z by addition;

Character according to the m sequence obtains again

$A^{H}A={\left[\begin{array}{cccc}2N& -2& \·\·\·& -2\\ -2& 2N& \·\·\·& -2\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ -2& -2& \·\·\·& 2N\end{array}\right]}_{L\×L}$

To A ^HA inverts, and obtains

${\left(A^{H}A\right)}^{-1}=\frac{1}{2(N+1)(N-L+1)}\left[\begin{array}{cccc}N-L+2& 1& \·\·\·& 1\\ 1& N-L+2& \·\·\·& 1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ 1& 1& \·\·\·& N-L+2\end{array}\right]$

$=\frac{1}{2(N+1)(N-L+1)}(\left[1\right]+(N-L+1)I)$

Wherein [1] expression all elements is 1 matrix entirely, I representation unit battle array.

Definition $a=\underset{i=1}{\overset{L-1}{\mathrm{\Σ}}}{z}_i,$ q＝(q ₁，q ₂，…，q _L-1) ^T，

According to q _i=a+ (N-L+1) z _i, i=1,2 ..., L-1 tries to achieve every value of q;

Last basis

$h=\frac{1}{2(N+1)(N-L+1)}(\left[\begin{array}{c}a\\ a\\ \·\\ \·\\ \·\\ a\end{array}\right]+(N-L+1)z)$ Obtain every value of channel parameter.

$=\frac{1}{2(N+1)(N-L+1)}q$

15, channel estimation system as claimed in claim 13 is characterized in that, described receiving system also comprises clock and carrier recovery unit, and described clock and carrier recovery unit are used for the described signal frame of channel is carried out clock recovery and carrier wave recovery.

16, channel estimation system as claimed in claim 13, it is characterized in that, described receiving system also comprises arithmetic element, described arithmetic element is used for the signal frame that carries out after clock recovery and carrier wave recover is carried out related operation, to find the original position of described pseudo random sequence, then described signal frame is sent to the channel estimating unit in the receiving system.

17, channel estimation system as claimed in claim 14, it is characterized in that, the length M of described Cyclic Prefix is more than or equal to the length L of channel, and the value that satisfies N-L+1 is 2 integer power, so that multiplying each other and invert by addition and shift operation realization matrix only.

18, channel estimation system as claimed in claim 13 is characterized in that, the duration of described burst communication is enough short, so that utilize channel information that channel estimating obtains effective in during burst communication.

Images